Throughout the ages, humans have devised ways to record, then examine, their thoughts and mental images on paper. People draw depictions of events, places, projects, and sets of objects. They chart scientific processes, demographics, weather conditions, and mechanical systems; and they diagram organizations, trade routes, music, and inventions. These drawings aid both the originator and subsequent audiences in picturing spatial or symbolic relationships. Paper drawings and maps are representations of the real world; but often the viewer must struggle to see only what is relevant amidst too much information. Transparent overlays can separate graphic information, but are cumbersome and restrictive. Reference documents such as catalogues, guide books, atlases, and encyclopedias gather together images and text descriptions; but the user must flip back and forth between pages to find, link, and compare information.
Today, computers are now being used to generate, compile, and retrieve such graphic records. However, they have not as yet enabled viewers to smoothly call forth sets of graphic data to inform and stimulate a sustained, multi-faceted, analytical thought process. For example, computer programs that employ graphics layering are currently used to generate illustrations (computer graphics), drawings of designs (CADD), and searchable and thematic geographic maps (GIS).
Computer graphics can be used to draw lines and shapes which may be organized into layers for overlapping and for showing and hiding before being output as a printed or digital illustration. CADD (Computer-Aided Design and Drafting) is used to create plans of products, vehicles, buildings, utility systems, and other three-dimensional objects. CADD employs layering technology to draft and show different views of a three-dimensional object. A GIS (Geographic Information System) plots data on a map with layers of points, lines, and polygons representing land features. Each GIS feature has an entry in a database with its name, a set of coordinates (which may be real-world latitude and longitude) for positioning the feature on the screen, and attribute data which may be shown in a separate window and queried to determine a set of features to be displayed.
These graphics-making programs are used primarily by technicians to produce a singular image for publication and, in limited ways, for interactive analysis. However, non-technicians—the public, executives, and experts in non-computer fields—cannot use them to easily access and manipulate selections of layered materials. Nor can they easily create or assemble their own sets of interactive, layered data.
GIS (Geographic Information Systems) has come furthest in adding interactivity to computer-generated images. Drop-down menus, graphics tools, and palettes are used for customized map production; this slow, unwieldy process is of very limited use for multi-faceted analysis. Networked GIS is used within corporate and government intranets to provide staff access to data displayed on geographic maps; it is also used on World Wide Web sites for the public to find a street address or the location of one particular facility or type of facility (see “Serving Maps on the Internet,” by Christian Harder, 1998, Environmental Systems Research Institute, Inc., Redlands, Calif., incorporated herein by reference as if restated in full.) In addition to searching for addresses and facilities, the interactive capabilities of current online GIS include zooming in and out and panning within a large digital map file.
A typical GIS often also has a large database with thematic information for places on the map. However, current GIS does not provide for seamless access to subset combinations from extensive data sets. A legend, table of contents, or key shows a small selection of themes and the symbols associated with subsets within each theme. In some cases, entire themes may be shown or hidden by clicking a check box next to the theme title on the legend; however, a particular subset within a theme cannot be shown or hidden. Current online GIS interfaces rely on the scrolling of palette scroll bars to extend the legend and to view selections for queries; this scrolling breaks the flow of data selection and severely limits the organization and amount of data that can be accessed. Sometimes scrolling is also required to view the entire map and each subsequent iteration, further breaking the flow of the analytical thought process. Annotational information, when available, is provided by going to a separate Web page or window; and thereby the material is not viewed smoothly, in direct association with the map feature.
Beyond layering software, another relevant computer technology is hypermedia. Hypermedia is the “linking” technology for instantly retrieving text, images, or sounds. Its “smart” graphics respond to commands such as mouse clicks to “hypertext” or to a “hot” symbol on the computer screen (often referred to as a button, object, icon, or imagemap).
From 1990 to 1993, the inventor developed a hypermedia map-making software program (“CityView/TownView HyperMapping—making maps and map libraries on your computer. A Journal of Demonstration Projects” and software manual, self-published, by Barbara L. Barros, Boston 1994, incorporated herein by reference). This program provided an interactive tutorial and tool with which novice computer-users could make their own map libraries for a study of their neighborhoods, cities, or towns. The software program used the first widely available hypermedia software development tools; APPLE COMPUTER INC.'s two-layered HYPERCARD. On the background layer of the map-making software was a base map either drawn by the user or imported as a scan or GIS-generated bitmap graphic. The user rendered sets of information on separate top layers using paint (raster or bitmap) graphics, text fields, and buttons which could link to new maps, paint layers, and text. The overlays were automatically indexed in a directory from which users could select layers to combine on a new, aggregated map. However, the layers were then fused together, so the map could not be used to aid a flowing analytical thought process.
The primary intent of the “CityView/TownView” map-making software was to enable civil servants and citizens to engage in exercises through which they could improve their awareness of local planning issues and opportunities. The software had several significant limitations: (1) the map features were not objects and could not have data or scripts attached to them, (2) map graphics could only be in black and white which reduced legibility and appeal, (3) the software was dependent on APPLE's MACINTOSH operating system having limited distribution, (4) users tended not to have or be able to afford base map data, and (5) the large amounts of graphic data generated could not be stored on most personal computers or shared easily with others.
CD-ROMs and the World Wide Web are, at this time, the two predominant means of delivering hypermedia. For organizing and displaying material, these current hypermedia use GUI (graphical user-interface) formats which are derivative. They draw from a combination of print publication equivalents, software document-creation conventions, and video presentation methods.
CD-ROMs were the first major application of hypermedia because of the vast storage capability for graphics and sound. These files are linked to create multi-dimensional games, training materials, and reference sources. Current reference CD-ROMS are organized to incrementally access graphic and text information on a single subject, but are not presently designed to enable comparison and to enhance prolonged, dynamic visual analysis.
The second major hypermedia application is the World Wide Web. Presently, the technology favors lengthy text over graphics, and graphics are commonly used as small link buttons and illustrations rather than as the core data set. While searchable GIS maps and ecommerce shopping sites are growing to be among the popular image-intensive applications on the Web, the quality of their interactivity and usefulness for visual analysis and comparison is extremely low. A complex quest is constrained by the slow speed of the Web's current infrastructure and its standard presentation methods.
For example, the Web's current accepted practice of presenting material is based on a page-to-page metaphor related to print magazines. The magazine-like vertical layout requires the scrolling of the page to accommodate the horizontal orientation of the computer screen. The page is made up of magazine-like article, illustration, and ad components. The eye roams from component to component, viewing each separately. Text is often lengthy. To obtain additional information, hypertext, symbols, drop-down menus, or query forms are clicked to call up an entirely new Web page. Often the new material is at a different Web site with a different format. The viewer attempts, with minimal success, to carry in the mind's eye the sequence of information, links, and pages. The viewer must build a mental model of the findings of their inquiry, instead of having this accomplished for them in the computer.